Web mapping and web apps for soil data

library(sf)
library(raster)
library(dplyr)

library(mapview)
library(RColorBrewer)

Datasets

For this demonstration, we are using the cookfarm dataset from the GSIF package. This dataset is a list with 6 somponents storing various measurements that have been done on the farm.

data('cookfarm', package = "GSIF")

names(cookfarm)

## [1] "readings"    "profiles"    "bdensity"    "grids"       "weather"    
## [6] "proj4string"

Vector data

The profiles component stores soil profile data as a data.frame. We will turn this into a simple featrure using the sf package:

profiles <- cookfarm$profiles %>% 
  st_as_sf(coords = c('Easting', 'Northing'), crs = cookfarm$proj4string)

# simple plot using the sf package
plot(profiles)

Raster data

The grids component is storing gridded data. We will reansform these into a RasterStack using the rasterFromXYZ function. This function requires the x and y columns to be the first column of the data.frame (from the left) – a good opportunity to use the select function from the dplyr package:

grids <- cookfarm$grids %>% 
  dplyr::select(x, y, DEM, TWI, Cook_fall_ECa, Cook_spr_ECa) %>%  # Re-order the columns, and select 4 interesting variables
  rasterFromXYZ(crs = cookfarm$proj4string) # The coordinate reference system is stored as one of the component in cookfarm

# simple plot using the raster package
plot(grids)

Interactive maps using mapview

Let’s start easy…

Vector data

# couldn't be easier to create a map!
mapview(profiles)

# Tweaking palettes 
pal_continuous <- colorRampPalette(brewer.pal(7, "BrBG"))
pal_categorical <- colorRampPalette(brewer.pal(9, "Set1"))

mapview(profiles, zcol = "TAXSUSDA", col.regions = pal_categorical, legend = TRUE)

# Tweaking backgrounds
mapview(profiles, zcol = "BLD", col.regions = pal_continuous, legend = TRUE, map.types = "Esri.WorldImagery")

# Burst to separate soil classes
mapview(profiles, zcol = "TAXSUSDA", col.regions = pal_categorical, legend = TRUE, burst = TRUE)

Raster data

# Plot grids
mapview(grids)

mapview(grids, col.regions = pal_continuous, na.color = NA,  legend = TRUE)

mapview(grids$TWI, col.regions = pal_continuous, na.color = NA, legend = TRUE)

Satellite data

# Raster specific functions
viewRGB(poppendorf)

viewRGB(poppendorf, 4, 3, 2)

Visualising projected data

plainview(grids$DEM)

Adding mapviews together

# Plot both together
mapview(grids$DEM, col.regions = pal_continuous, legend = TRUE) + mapview(profiles, zcol = "TAXSUSDA", col.regions = pal_categorical)

Advanced tools

Sync’ing maps

# Syncing several maps
m1 <- mapview(grids$DEM) 
m2 <- mapview(grids$TWI) 
m3 <- mapview(grids$Cook_fall_ECa)

sync(m1, m2, m3, ncol = 2, sync.cursor = TRUE)

Slideview

img1 <- poppendorf[[1]]
img2 <- poppendorf[[5]]

slideview(
  img1, 
  img2,
  label1 = "Poppendorf-Layer-1",
  label2 = "Poppendorf-Layer-2",
  legend = TRUE
)

Popups

mapview(
  profiles, 
  popup = popupImage('https://www.vcard.wur.nl/WebServices/GetMedia.ashx?id=37263')
)

Advanced mapview

library(xts)

## Loading required package: zoo

## 
## Attaching package: 'zoo'

## The following objects are masked from 'package:base':
## 
##     as.Date, as.Date.numeric

## 
## Attaching package: 'xts'

## The following objects are masked from 'package:dplyr':
## 
##     first, last

library(dygraphs)

profiles$SOURCEID <- as.character(profiles$SOURCEID)

records <- cookfarm$readings
records$SOURCEID <- as.character(records$SOURCEID)

ids <- unique(records$SOURCEID)

# Subset sensors
ids <- sample(ids, size = 5) 
  
idx_sensors <- which(profiles$SOURCEID %in% ids)
sensors <- profiles[idx_sensors,]

make_ts <- function(id) {
  
  records %>% 
    filter(SOURCEID == id) %>% 
    dplyr::select(-SOURCEID) %>%
    dplyr::select(Date, ends_with('VW')) %>% 
    xts(.$Date)
}

make_dygraph <- function(id){
  ts <- make_ts(id)
  dygraph(ts)
} 

l_graphs <- lapply(
  ids,
  make_dygraph
)

make_dygraph(ids[1])

mapview(sensors, popup = popupGraph(graphs = l_graphs, width = 300, height = 300))